The Coming Negative NAO: Implications

[donsutherland1]

Over the past few days, the GFS ensembles have locked in on the idea of the development of a negative NAO. Now, there is a strong consensus among the ensemble members that the NAO will head to -2.00 by March 1:



At the same time, the NCEP ensemble 500 mb mean anomalies suggest a trough near or off the East Coast that is 3 or more standard deviations below normal:



Already, the NAO is falling and it will probably go negative either tomorrow or Wednesday:

2/16 +0.901
2/17 +0.702
2/18 +0.490
2/19 +0.215
2/20 +0.150

The implication would be a generally colder-than-normal pattern in the eastern United States. The historical data (1950-2004) for the 2/20-29 period supports such a proposition.

Mean Temperatures for Given NAO Ranges:

Boston:
NAO > 0: 35.6°
NAO -0.99 to 0: 32.1°
NAO -1.99 to -1.00: 33.1°

New York City:
NAO > 0: 39.7°
NAO -0.99 to 0: 35.7°
NAO -1.99 to -1.00: 36.1°

Philadelphia:
NAO > 0: 39.8°
NAO -0.99 to 0: 36.1°
NAO -1.99 to -1.00: 36.9°

Washington, DC (DCA):
NAO > 0: 43.7°
NAO -0.99 to 0: 40.2°
NAO -1.99 to -1.00: 40.8°

Percentage of Days with Highs < 40°:

Boston:
NAO > 0: 18%
NAO -0.99 to 0: 27%
NAO -1.99 to -1.00: 25%

New York City:
NAO > 0: 23%
NAO -0.99 to 0: 35%
NAO -1.99 to -1.00: 41%

Philadelphia:
NAO > 0: 20%
NAO -0.99 to 0: 35%
NAO -1.99 to -1.00: 33%

Washington, DC (DCA):
NAO > 0: 10%
NAO -0.99 to 0: 18%
NAO -1.99 to -1.00: 20%

In addition, there are also some snowfall implications. The Mid-Atlantic region sees a somewhat higher likelihood of accumulating snow.

Percentage of Days that receive 0.1" or More Snowfall:

i]Boston:[/i]
NAO > 0: 17%
NAO -0.99 to 0: 20%
NAO -1.99 to -1.00: 19%

New York City:
NAO > 0: 5%
NAO -0.99 to 0: 10%
NAO -1.99 to -1.00: 13%

Philadelphia:
NAO > 0: 6%
NAO -0.99 to 0: 11%
NAO -1.99 to -1.00: 7%

Washington, DC (DCA):
NAO > 0: 3%
NAO -0.99 to 0: 7%
NAO -1.99 to -1.00: 7%

It should be noted that the magnitude of change that is forecast in the NAO's regime change to negative has sometimes coincided with important storms in the East per Heather Archambault's research.

As a result of the developing block, my idea (see my February 16-28 expectations) of moderation to near normal in the 2/23-28 period will probably see an attempted moderation turned back in the closing days of the month. In fact, the NCEP ensemble mean suggests the possibility of a very cold air mass to close out the month and that suggestion appears quite reasonable.

Moreover, with such a deep trough developing to coincide with the growing block, the Mid-Atlantic and New England states will need to be wary of the possibility of at least one or two shots of accumulating snow prior to the end of February. The greatest risk of a moderate to large storm would probably lie in the closing days of the month.

Finally, should such blockiness persist through the first half of March, that would also imply a colder than normal first half of the month. At this time, given the ENSO analogs (ENSO regional anomalies and MEI), past tendencies for weak La Niña/PDO+/QBO East seasons, I believe March will probably prove both cooler than normal and snowier than normal from Washington, DC to Boston. This does not mean that there won't be warm days, as I do believe that the weaker and probably less persistent blocking episode relative to last year's will allow for more warmth.

Interestingly enough, the two analogs per the ENSO regional anomalies and MEI that saw February snowfall of 8" or more in Washington, DC (DCA) and 15" or more at both New York City and Boston were 1966-67 and 1995-96. The small sample size suggests a measure of caution is in order. However, the larger historical patterns (based on seasonal snowfall to date) and La Niña-PDO+ winters suggests a good possibility for above normal snowfall in March.

For now, a block is forming and that should translate into a colder outcome to close February and possibly opportunities for additional snowfall, as well.

[Hybridstorm_November2001]

Correct me if I'm wrong, but didn't a Negative NAO, and a Weakening La
Nina, also play a role in the wet and stormy Spring and First half of Summer we had in the East in 1996?

[donsutherland1]

It did, Hybridstorm_November2001.

Winter 1995-96 featured a weak La Niña/PDO+ setup. It was historically snowy in parts of the East with many cities establishing all-time seasonal snowfall records.

[Hybridstorm_November2001]

Also we had three tropical systems; Bertha, Fran and Josephine effect the region that summer. Not to mention my area received some wind and rain from the NW fringe of Hortense, and of course the treat and eventual near miss presented to the Mid-Atlantic and New England by Edouard.

[terstorm1012]

Also we had three tropical systems; Bertha, Fran and Josephine effect the region that summer. Not to mention my area received some wind and rain from the NW fringe of Hortense, and of course the treat and eventual near miss presented to the Mid-Atlantic and New England by Edouard.

I tell you what Edouard was scary . . . for a day or two it looked like it was going to come right up Delaware Bay at Philadelphia. It woke some people up there, but eventually he turned and went out to sea.

As for the Negative NAO, a coworker (there's a group of us who are weather nerds at work) mentioned it was going to be cold through the beginning of March, to which I groaned, because I'm ready for spring. Been stuck in the house too long...

[Hybridstorm_November2001]

Same here, but if it is anything like 1996, Spring could be quite chilly and late in the NE

[Cowhide]

Well wasn't most of the cold arctic air stationed on the opposite side of the northern hemisphere for most of December and January? And maybe it's making up for the very warm winter come March?

Also. I'd like to know: Why does it get colder in Siberia than in the North Pole, does it have anything to do with the sun angle?

[Hybridstorm_November2001]

Land heats up and cools down faster than water, there isn't much land at the North Pole itself. Another fact is that there are no currents over land areas to move around hot and cold like in the Oceans.

[ohiostorm]

If I remember correctly. We had alot of snow here in Ohio during the late parts of 96. Could this be a repeat?

[luvwinter]

I remember that year Ohio Storm. I live in Dayton, Ohio and it seemed like we got snow a couple of times a week from Thanksgiving until atleast the end of January. It was the best winter in a long time.

[donsutherland1]

Just a quick note...

The NAO has now fallen below 0. The 2/22 reading was -0.115. It will be interesting to see how strong this block actually grows, how long it lasts, and whether it can deliver, both in terms of cold and opportunities for snowfall.

[pawxguy]

I hope you are all reading between the lines/listening with both ears, etc because the next 10 days will be an incredible buildup to what MAY happen next weekend along the east coast. I, for one, will enjoy the hype leading up to our next potential event.

[Stephanie]

I hope you are all reading between the lines/listening with both ears, etc because the next 10 days will be an incredible buildup to what MAY happen next weekend along the east coast. I, for one, will enjoy the hype leading up to our next potential event.

You're talking about the local mets, correct????

[pawxguy]

I hope you are all reading between the lines/listening with both ears, etc because the next 10 days will be an incredible buildup to what MAY happen next weekend along the east coast. I, for one, will enjoy the hype leading up to our next potential event.

You're talking about the local mets, correct????

I'm not speaking 'up' to them; I am merely speaking to everyone else on the board Stephanie. The local mets are better off not even mentioning such a discussion to their audience until THIS weekend is over.

[donsutherland1]

Pawxguy,

While I have good and growing confidence in the idea that the Mid-Atlantic and southern New England regions will likely see above normal snowfall for March, it's a little more difficult to suggest whether or not there will be a second major to historic snowstorm this winter. Things could well evolve like last winter's superblock where fairly frequent small to moderate snowfalls led to high totals from late February into mid-March. On the other hand, there may well be an opportunity for big event. I'm open to either possibility.

[Ivanhater]

don, i just wanted to say, thank you for taking your time to post your thought on tropical to winter weather...i know it takes a while to come up with your findings and i appreciate you posting them and explaining them to us...looking foreword to your thoughts during this fast approaching hurricane season

[boca]

I second ivanhater's thoughts on this thanks Don.

[donsutherland1]

Some morning thoughts...

1) The block is now building as per the latest NAO readings:

2/21: +0.098
2/22: -0.115
2/23: -0.493

The GFS ensembles continue to show the NAO bottoming out near -2.000 around March 1 +/- 1 day. Afterward, generally after 3/5, the NAO is forecast to return to positive readings. This timing fits well with past experience with blocks of similar magnitude.

Excerpts from 2/21:

Longer-term, I like the idea of a weakening negative NAO, particularly after March 1...and possibly its going > 0 in the 3/5-12 timeframe...

I looked at past blocks that bottomed out in the -1.50 to -2.99 range while forming 2/15-3/5 and found the following:

∙ Median duration: 14 days
∙ Mean duration: 15.4 days
∙ Shortest: 7 days (2/26-3/4/1998)
∙ Longest: 23 days (2/26-3/20/1954)


2) This block will likely retrograde to somewhere roughly in the vicinity of the Hudson Bay within the next 8-12 days. The progression of the block on the NCEP ensemble 500 mb anomalies illustrates this process quite well.

168 Hours:


240 Hours:


One should bear in mind that some changes could occur in future runs, but I believe the broad idea of a retrograding block is reasonable.

3) Near-term snowfall prospects still exist, but odds of a big storm remain low. The passage of the Arctic front could generate some minor accumulating snow. However, low but not zero odds of coastal development that would pose the greatest threat to eastern Long Island and eastern New England and possibly Philadelphia north and eastward appear to have fallen in recent days.

4) Longer-term snowfall prospects from the episode of developing blocking: The question remains as to whether the blocking episode underway will produce a big East Coast snowstorm before it fades.

Right now, two scenarios appear possible:

∙ Somewhat frequent light to moderate snows for the Mid-Atlantic to New England region (particularly from DCA northward).
∙ A big major to perhaps historic snowstorm.

As noted, early on the prospects for a big storm (see point #3) appear somewhat low and perhaps a little lower than a few days ago. They're not zero but odds are probably against such a storm.

This leaves the 3/1-10 period under consideration. Historically, odds strongly suggest that DCA to BOS have not seen their last accumulating snows of the season. But such odds do not necessarily suggest a major to historic storm. Certainly, a number of seasons have witnessed multiple Kocin-Uccellini storms (1957-58, 1960-61, 1968-69, 1977-78, and 1986-87).

Since 1950, 1 such storm occurred in the 2/20-3/10 timeframe at a time when the block had retrograded to a position near the Hudson Bay: the long-lived 2/22-28/1969 system. That system focused its heaviest snows on portions of New England, where incredible amounts fell.

500 mb Anomalies: 2/19/1969:


500 mb Anomalies: 2/22/1969:


If one compares the above charts with the NCEP anomalies posted earlier, one finds at least a few very general similarities:

∙ A retrograding block (but the forecast block is somewhat to the north of that for the 1969 storm)
∙ Troughiness near or off the U.S. West Coast
∙ Troughiness in eastern North America (but the forecast trough is farther to the northwest than for the 1969 storm)

One also finds a handful of smaller but still meaningful storms in which a block had retrograded to the general vicinity of the Hudson Bay region in the larger 2/20-3/20 period:

March 14, 1958
February 25, 1966
March 6, 1990
March 5, 2001

The synoptic setups differed quite markedly for each of these cases. This diversity of setups argues that there is probably wider latitude for a more moderate system than a major to historic one.

Given the historic data that argues for additional snowfalls and the differences between the forecast pattern and that for the 1969 storm, for now, I lean toward the more frequent but more moderate storms for the 3/1-10 period. This does not mean that I don't believe that a larger one cannot occur, just that I'd like to see more data and more support for such a storm before arguing for its likelihood. Such storms are not common, so it makes sense to start from a reasonably conservative perspective on that end and then make adjustments from there should the data grow more conducive for such a storm. Even then, I'd argue that a major storm would probably be far more likely than a historic one given the current data and climatological odds.

[donsutherland1]

Thanks Boca and Ivanhater.

Best wishes.

[MiamiensisWx]

How do you think this will effect the upcoming hurricane season, donsutherland1?